1. Most of the tubes would need some bowing anyway for curves, since they can't find a perfect straight path from LA to SF 2. Tests show that tubes where L >>

1. Most of the tubes would need some bowing anyway for curves, since they can't find a perfect straight path from LA to SF 2. Tests show that tubes where L >> D, thermal expansion is linear along the length. If you restrained both ends you'd get the buckling in the interior sections that you describe. That might even be a bigger problem though because it would affect the path geometry and alter the attitude control requirements of the pods. I think the wrinkling in the tube skin would also be more dangerous than just letting it stretch.